32.1DuncanTBMDedited © Mike Duncan. Informal Logic, Vol. 32, No. 1, (2012), pp. 83-97. The Curious Silence of the Dog and Paul of Tarsus: Revisiting The Argument from Silence MIKE DUNCAN University of Houston Downtown One Main Street Houston, Texas 77002 duncanm@uhd.edu Abstract: In this essay I propose an interpretative and explanatory struc- ture for the so-called argumentum ex silento, or argument from silence (henceforth referred to as the AFS). To this end, I explore two examples, namely, Sherlock Holmes’s oft- quoted notice of the “curious inci- dent of the dog in the night-time” from Arthur Conan Doyle’s short story “Silver Blaze,” and the histori- cal question of Paul of Tarsus’s si- lence on biographical details of the historical Jesus. Through these cas- es, I conclude that the AFS serves as a dialogical topos best evaluated and understood through the perceived authority of the arguer and the wil- lingness of the audience to accept that authority, due to the “curious” nature of the negative evidence that the argument employs. 
 Résumé: Dans cet essai, je propose une structure interprétative et expli- cative pour le soi-disant argumen- tum ex silento, ou un argument qui fait appel au silence (ci-après dé- nommé l 'AFS). À cette fin, j'explore deux exemples, à savoir, une remar- que souvent citée de Sherlock Holmes, "l'incident curieux du chien [silencieux] dans la nuit" tiré du conte "Silver Blaze" de Arthur Co- nan Doyle et la question historique du silence de Paul de Tarse sur les détails biographiques du Jésus his- torique. Grâce à ces cas, je conclus que l'AFS sert d’un topos dialogique mieux évalué et compris à travers l'autorité perçue de la personne qui avance l’argument et à travers la volonté de l'auditoire d’accepter cette autorité, en raison de la "cu- rieuse" nature de la preuve négative que l'argument emploie. Keywords: acceptability, argument from silence, authority, dialectic, exper- tise, fallacy, rhetoric. 1. Past and current thought Scholarly examinations of the Argument From Silence (AFS) are extremely rare; when existent, it is typically treated as a fal- lacy.1 However, of the two indispensable theoretical texts on 



























































 1 Lange, writing in 1966, mines a smattering of texts on historiography, ranging from 20-80 years old at his writing (288); his piece, as well as the Mike Duncan 84 fallacies in the last 2500 years, Aristotle’s Sophistical Refuta- tions and Hamblin’s Fallacies, neither explicitly mentions an argument from silence,2 and nor is it mentioned in the develop- ing pragma-dialectical standards (van Eemeren et al). The practical dimensions of the AFS vary, save the con- stant that a lack of evidence is presented as evidence that some- thing is possible, what Stephens calls “absent evidence reason- ing” (56). Sometimes, the claim of possibility is extended to “probable” or “highly probable,” the latter examples tending to attract a charge of logical impossibility. Quite often, though, even the more modest claim of “possible” receives the same log- ic critique. Behind this critique is the assumption that considering ap- parent lack of evidence as evidence of actual lack of evidence is to go too far, which in turn raises the question of how any lack of evidence can ever be established as meaningful. Advice con- sidering AFS use is universally tinged with caution. This par- ticular critique is widespread enough that it is difficult to locate a positive use of the phrase “argument from silence” in the lit- erature of any field—an observation that, itself, constitutes an AFS. John Lange’s exploration of the AFS in 1966 notes its worth as a historical argument and its critiqued status in research methodologies. Lange rejects an insistence that the AFS must be conclusive in all cases (290), realizing that this would throw out all kinds of historical arguments that are commonly perceived as reasonable. Instead, he cannibalizes a relatively relaxed set of preconditions from historiographical texts (which presuppose that the dating/existence of a historical event is under question): 1. There is a document, D, or a functional equivalent the- reof, which is extant, in which the event, E, is not mentioned. 2. It was the intention of the author of D to enumerate exhaustively all members of the class of events of which E is supposed to be a member. 3. E must be such that, if it had occurred, the author of D could not have overlooked it (290-291). 































































































































 brief discussion in Brickman’s 1949 Guide to Research in Educational History (170-172) is very dependent on Gilbert Garraghan’s 1946 A Guide to Historical Method. 2 This is a pretty strong example of an AFS in action. If Aristotle and Hamblin did not mention the AFS—especially Hamblin—they very likely did not think it was a fallacy, or that it was worthy of discussion in context of fallacy. Both discuss the argument from ignorance, though, which has some similarities, as Walton discusses (374).
 Revisiting the Argument from Silence 
 85 Lange concludes that “specific instances” of the AFS must be evaluated individually; there are no good general dialectical rules to follow, save perhaps expertise with the evidence offered (290-291). Dismissing all instances is counterproductive, as AFS is never logically or rationally conclusive, only suggestive (301). Lange’s singular study in important, but he does not pro- vide an extended example of an AFS to show how it interacts with other competing arguments on the same question, or pro- vide guidelines for assessing its strength relative to those other competing arguments. He does call for such guidelines, ideally cooperatively developed by a “historian and some probability logician,” but does not offer any (301). He does provide several short AFS examples taken from methodological works (289), but he does not explore any of them. Douglas Walton briefly discusses the AFS as a variant of argumentum ad ignorantiam or appeal to ignorance (371). He compares it to the standard of negative evidence in scientific re- search, which is typically devalued when compared to the power of positive evidence (372). Walton does not explore the AFS much further, but he seems to hold that like in other nonfalla- cious uses of argumentum ad ignorantium, that the AFS, at least when nonfallacious, is based on knowledge—albeit of a nega- tive kind—rather than ignorance (373). For example, Walton outlines several instances of apparently nonfallaicous arguments from ignorance, including a lengthy textbook example where the FBI fails to find any evidence showing than a “Mr. X” is a communist, and therefore concludes that Mr. X is probably not a communist, As this instance “combines ignorance with know- ledge,” through the extensiveness of the FBI’s search, the result- ing argument is more persuasive (371). The special nature of this negative knowledge, however, is not explored further. Most recently, Christopher Stephens has offered a Baye- sian approach to absent evidence reasoning, suggesting that the epistemological motto “absence of evidence is not evidence of absence” does not hold true in cases where evidence has been actively searched for (such as the FBI example featured in Wal- ton) as the results of such a search constitute evidence that shifts the probability of a question, given an informal logic framework (56). While Stephens clarifies to an extent the nature of the knowledge produced by observing silence in such arguments, and confirms the nonfallacious nature of the motto in such special cases, he does not examine the effect of the arguer(s) and or the audience(s) on the argument, both of which are crucial to evaluating its informal structure and strength. Reviewing Lange, Walton, and Stephens suggests that more could be learned about the AFS by reviewing several AFS Mike Duncan 86 examples at length to determine, first, how negative knowl- edge operates differently than other kinds of offered evidence (in other words, what makes the AFS different from other argu- ments) and second, how the rhetorical relationship between ar- guer, audience, and context affects use and reception of the AFS. As such, I offer the following two case studies—both well-known, though one is fictional in nature, and the other his- torical by nature—as fodder for discussion. 2. The curiously silent dog In Arthur Conan Doyle’s short story “Silver Blaze,” Sherlock Holmes is investigating a murder that involves a stolen horse, Silver Blaze, who has gone missing from a stable. Upon realiz- ing that the stable’s resident dog did not bark on the night of the crime—Holmes is not told this, but infers it through no one mentioning a bark3—he further infers that the horse was stolen by someone the dog knew well.4 His reasoning might be broken down thusly: The dog would have barked at any strangers in the stables on the night in question. The dog did not bark on the night in question. Therefore, the dog probably did not encounter a stranger on the night in question. As such, Silver Blaze probably was not stolen by a stran- ger to the stables. This example introduces the main limitation of the AFS. It can only ever establish probability, though the prima facie case pro- duced by a strong AFS is often valuable. In this case, the dog’s lack of alarm alerts Holmes to the possibility and probability that the horse was stolen by someone the dog knew. It is critical to note that the following argument, while ini- tially seeming very similar, is quite different: The dog would have barked at all strangers in the stables on the night in question. The dog barked on the night in question. 



























































 3 In a way, this is a double-layered AFS; Holmes takes the absence of notification of a bark as the absence of a bark.
 4 Holmes’s explanation of the significance of the dog’s silence (Doyle 280) suggests that he made the inference well before his initial and well-quoted comment on it (Doyle 277). Revisiting the Argument from Silence 
 87 Therefore, the dog probably encountered a stranger that night. As such, Silver Blaze probably was stolen by a stranger to the stables. The difference lies in how a definite event (the barking) can be easily deemed evidence, while the lack of an event (the absence of barking) is more difficult to grasp intuitively as a premise. When Holmes calls the silence a “curious incident,” Doyle plays with the notion of what an “incident” is; how can the absence of something be an event? Cheryl Glenn has shown powerfully how silence itself can be a rhetorical act (2); the dog’s silence on any given night, then, could be seen as testimony that there have been no strang- ers nearby, a tacit “all clear” signal, just as much as frequent barking means the opposite. That the guard is not human makes little difference; substitute a night watchman who shouts in alarm and the reasoning remains. In this sense, a silence or ab- sence that violates previous notions of usual behavior can ac- quire a warrant and passes into the realm of “incident,” and sub- sequently, evidence. However, this transformation has special limitations, in that the perception of a gap in a pattern—in behavior, in the case of the dog—does not necessitate that the gap exists, or even that the gap, if existent, means anything or has a discernable cause, as Hume’s example of the missing blue shade from his Enquiry suggests (12-13). Doyle plays with this special limitation in an- other Holmes story, where a man that Holmes has been follow- ing claims to have seen no one following him, and Holmes re- marks dryly, “That is what you may expect to see when I follow you.”5 The absence of evidence becomes evidence—but only through Holmes’s interpretative act, and only if his judgment is accepted by the audience. There is, perhaps, an even more curious and telling silence in “Silver Blaze,” however, than that of the dog’s: no one in the story criticizes the logic behind Holmes’s snap inference. The premises are far from ironclad: the dog could have been asleep, or drugged (like the stableboy was), or placated with food, or even made its first exception for a friendly stranger. Likewise, the witnesses could have been mistaken about not hearing a bark through distraction or inattention—the dog may have actually barked quite a bit—or the witnesses could have been lying. Holmes’ establishment of the reliability of the dog and the wit- nesses is simply assumed; there is no indication that he has ex- 



























































 5 From “The Adventure of the Devil’s Foot” (Doyle, 521). Mike Duncan 88 plored any of these possible alternate explanations. As such, his confidence in the evidence would be fallacious under the ex- ploratory standard proposed by Stephens, as the silence in ques- tion is not sufficiently observed and confirmed. However, this curious silence by potential, skilled crit- ics—Dr. Watson and the police inspector—suggests that Holmes has so established himself as an expert in regard to the premises of the argument that they do not feel the need to ques- tion its relevance. Doyle dramatizes Holmes’s statement by placing it at the very end of a scene (277), suggesting Holmes’s audience was left speechless by either assuming Holmes as ex- pert had made a connection they were unable to make them- selves, or by realizing, perhaps dimly, the importance of his ob- servation. In either case, Holmes assigns meaning to the dog’s silence by the power of his authority as a well-respected consult- ing detective, who knows what kind of canine silences can con- stitute warranted evidence for claims—and Watson and the po- lice accept this authority regardless of their own self-perceived levels of understanding. It is also important to note the special function of this par- ticular AFS in the larger scheme of Holmes’s eventual solving of the story’s mystery. Holmes’s observation about the dog’s silence functions a high-probability guess, a conductive step- ping-stone that leads him to more solid argumentative ground where he can gather positive evidence that is more easily di- gestible. The dog’s silence does not prove that Jonathan Straker, the actual thief, stole Silver Blaze, but making the assumption allows Holmes to construct theories about Straker’s later death that would have been difficult to do so otherwise without con- ceiving of him as the horse thief. The AFS thus serves as an ex- ploratory tool for Holmes, allowing him access to stronger evi- dence that eventually confirms rather than suggests Straker’s theft of the horse. “Silver Blaze,” then, establishes two important points about the AFS. First, the highly probabilistic nature of the evi- dence shifts much of the burden of proof from the evidence it- self to the strength of the arguer’s ethos as an expert on such evidence. The comment of some random passerby on silence of the dog might or might not have led to further investigation (or, even, an inquiry into the reliability of the dog and the witness); a comment by Sherlock Holmes, however, focuses the attentions of the police and even the highly skeptical Colonel Ross. Sec- ond, the argument itself is a conductive stepping-stone on Holmes’s way to a larger, more elaborative and conventionally supported argument about who stole the horse and who killed Straker. The silence of the dog is no longer questionable nega- Revisiting the Argument from Silence 
 89 tive evidence, but a necessary conductive link in Holmes’s self-styled “chain of reasoning,” where “one true inference in- variably suggests others” (Doyle 279-280). His abductive rea- soning style (Diderjean et al 15), which begins with an existing state of affairs and reasons backwards to find causes, would be impossible without such a tool. 3. The curiously silent Paul of Tarsus I have offered the Holmes example for accessibility, but the AFS is also present in difficult historical questions. Doyle’s Holmes, as observed before, enjoyed an absence of serious de- tractors in his opinion of the dog’s silence, and he had consider- able access to physical evidence. However, in the case of Paul of Tarsus’s silence, there is no such uniformity of opinion, and no plethora of evidence. Pauline scholarship in biblical studies has long wrestled with the question of how to explain why the known authentic letters of Paul of Tarsus present in the New Testament contain so little material concerning the life of a historical Jesus. Paul references no parables, no speeches, no birth narrative, no virgin birth, no John the Baptist, no miracle-working, no exorcisms, no trial, no scourging, and no accounts of most of the twelve apos- tles, especially Judas, among many other silences—some seem- ingly conspicuous, others not—that are too numerous to list in- dividually here. The instances of relatively concrete knowledge of material that can also be found in the canonical gospels are scarce, besides the crucifixion and a few apparent teachings.6 Given these circumstances, it is possible to argue that Paul did not know of much of the material concerning Jesus’s life in the canonical Christian gospels when writing his letters in the 50’s and early 60’s CE, because if he had known about this ma- terial, he would have mentioned more of it in his letters. The ar- gument can be reduced to the following: 



























































 6
Paul mentions Jesus has Davidic descent (Rom 1.1-3, Rom 9.3-5), that he was handed over during a possibly liturgical meal (1 Cor 11:23-27), and that he gave a directive to live from the gospel (1 Cor 9:14), as well as another directives on marriage and divorce (1 Cor 7:10-11). Furthermore, Paul offers a compact set of six accounts of post-resurrection appearances of Jesus, in- cluding his (Paul’s) own revelation (1 Cor 15:3-5), and a few “nests” (Walter 56) of Jesus-like sayings (Rom 12:14-21, 1 Cor 4.11-13). The bulk of this material, save a few questionable parallels, is in two of the seven known au- thentic letters. 
 Mike Duncan 90 Paul, writing earlier than the gospel authors, does not mention many key aspects of the life of Jesus in the gospels in his letters. The gospel authors, writing later than Paul, mention a great deal of material about the life of Jesus in their works. Therefore, Paul did not know much about the life of Jesus. This argument is a non sequitor; Paul does not have to write about Jesus’s life to know about it, and there is no necessary connection between the knowledge possessed by the gospel au- thors and Paul’s knowledge. Pointing out the logical error, though, does not address the core assumption behind this argument and behind all AFS usage in historiography; namely, the absence of many key details of Jesus’s life in Paul’s letters raises the possibility, however small, for Paul not knowing about much of Jesus’s life, just as the dog’s lack of barking in “Silver Blaze” caused Holmes to won- der if the dog knew the horse thief. In other words, if the two premises are accepted, then one must acknowledge that it is pos- sible, however unlikely, that Paul does not know about much of the gospel material. This is not how the AFS is usually stated, though, so here is a sturdier version: Paul, writing earlier than the gospel authors, does not mention many key aspects of the life of Jesus in the gospels in his letters. The gospel authors, writing later than Paul, mention a great deal of material about the life of Jesus in their works. Therefore, other evidence notwithstanding,7 it is possible that Paul did not know much about the life of Jesus. The carefully circumspect conclusion of this version is harder to challenge; it would require presenting evidence that Paul actu- ally did know of Jesus’s life outside of his letters, which would render the first premise true enough, but nothing more than a curiosity. A third premise, however, changes the argument sig- nificantly: Paul wrote extensively on Jesus and would have men- tioned major events in Jesus’s life in his letters if he had known about them. 



























































 7 I should note here that this form of the AFS is not some variant of modus tollens, as witnessed by “other evidence notwithstanding.” Revisiting the Argument from Silence 
 91 Paul, writing earlier than the gospel authors, does not mention many key aspects of the life of Jesus in the gospels in his letters. The gospel authors, writing later than Paul, mention a great deal of material about the life of Jesus in their works. Therefore, other evidence notwithstanding, it is possible that Paul did not know much about the life of Jesus. The new supporting premise depends on textual expertise ren- dering a judgment of probability, which lends the argument both strength and weakness; noting Paul’s focus on Jesus helps the case, but it is difficult to qualify what “would have mentioned” amounts to. Rewording it as “probably would have mentioned” or even “more likely than not would have mentioned” does not help matters; the probability under question is unquantifiable and can only qualitatively judged, like the silence of the dog in “Silver Blaze,” by a human expert who is familiar with the long- term reliability of similar evidence. Stephens attempts to evaluate similar values with Baye- sian analysis, but the Bayes theorem cannot tell us the level of increased probability the audience should take into consideration (62) as it cannot generate numerical values without numbers to input. As such, recalling the centrality of expertise, the AFS is not complete without the following rhetor-centric addition: Paul wrote extensively on Jesus and would have men- tioned major events in Jesus’s life in his letters if he had known about them. Paul, writing earlier than the gospel authors, does not mention many key aspects of the life of Jesus in the gospels in his letters. The gospel authors, writing later than Paul, mention a great deal of material about the life of Jesus in their works. Therefore, other evidence notwithstanding, in my profes- sional opinion as an expert on early Christian history and Pauline discourse, it is probable that Paul did not know much about the life of Jesus. The special evidential transformation that powers Holmes’s pro- fessional inference about the horse theft also rests behind this version of the AFS; namely, some silences are “deafening” enough to an expert that they can establish probability, not just possibility, of a given state (“likely” rather than “unlikely”) and that same probability then can become, potentially, acceptable to Mike Duncan 92 an audience. Holmes’s argument, too, can be restated in a similar fashion: The dog would have barked at all strangers in the stables on the night in question. The dog did not bark on the night in question. Therefore, other evidence notwithstanding, in my profes- sional opinion as a counseling detective familiar with canine behaviors, the dog probably did not encounter a stranger that night. As such, Silver Blaze was probably not stolen by a stran- ger to the stables. 4. Acceptability, expertise, and mystery-solving Given Johnson and Blair’s three standards for evaluating argu- ments—relevance, acceptability, and sufficiency (55), it is the acceptability standard where an AFS encounters difficulty. Relevance and sufficiency are handled by the judgment of the author/expert, but acceptability rests on that same ethos, espe- cially if the audience lacks comparable expertise that would al- low independent judgment of the evidence. Furthermore, even given comparable expertise, in the Pauline case, no level of per- suasive ethos may be enough to sway opinion. This ongoing requirement of an expert to champion an AFS suggests that the AFS is particularly dependent on author- ity; the evaluator must trust the arguer to have selected a silence that is relevant, meaningful, and warranted, and that trust must come from some judgment of the arguer’s apparent expertise as much as the actual lack of evidence. Fortunately for Holmes, the expertise required to show that a dog known to bark at strangers will almost certainly bark at a strange person in its demesnes is not terribly demanding to anyone familiar with dogs. However, there are many experts (and non-experts) on Pauline discourse that would not interpret Paul’s silence on Jesus life as a strong suggestion he knew little about the subject. As such, the AFS must also deal with argu- ments that counter its claims. A possible counterargument to an AFS concerning Paul’s silence could be stated thusly: Speaking as an expert on Pauline discourse, I hold that Paul probably knew all, most, or some of the material that we have in the canonical gospels. He did not mention this material much in his letters because he had no need or de- Revisiting the Argument from Silence 
 93 sire to do so, given he was strongly focused on a post- earthly Jesus, and considered such material either A) triv- ial, well known, and not worthy of mention, B) inauthentic or corrupted compared to his personal revelation, C) authentic, but not agreeable with his own teachings, which he wanted to stress, or D) some or all of the above. Comparing this counterargument to the AFS stated earlier re- veals some interesting structural details. First, a multitude of possible Pauls is postulated to counteract the single Paul of the AFS. While the AFS forms what Walter refers to as a “minimal hypothesis” on this question (53)—it does not need to assume more than what evidence is available at present—any counterar- gument must sort through a plethora of available options. It is possible to restate this counterargument using only one option (for example, option C), but that would forgo the usefulness of pointing out all of the other plausible options that the AFS has failed to consider. Their sheer numbers would seemingly over- whelm the AFS’s lone claim and allow for Paul to exist in an ambiguous state that is in many ways more useful for Pauline scholarship, with its diversity of opinions.8 Second, the focus of the counterargument, like that of the matching AFS, concerns the meaning of Paul’s silence. This de- pends on an understanding of who Paul was, and points toward the larger question of how the Pauline letters are to be inter- preted at all. If, for example, Paul is a man who received a pri- vate revelatory vision of Jesus (Gal 1:11–16) that he holds high- er than any message, myth, or narrative account coming out of the Jerusalem church, scholars must endeavor to read the letters (and the silence) in light of this ethical claim. If he was a prag- matist who balanced his revelation against various traditions or myths about Jesus that he encountered in his many travels about the Mediterranean (he was not the only roaming Christian missionary in the 1st century CE), scholars must interpret ac- cordingly. If he was a controlling personality who would use any authoritative source of persuasion to keep the small 



























































 8 This complicates the debate over the hypothetical Q sayings source, which is held to predate Paul’s letters, and contains much Jesus material. Confi- dence in the existence of Q allows the question of Paul’s knowledge of Jesus to be neatly sidestepped for many, for if Jesus sayings can be dated before Paul, then the need to defend Paul’s knowledge of Jesus is considerably dampened, for a Jesus tradition is safely preserved elsewhere. If Q is ephem- eral, however, then the question of Paul’s knowledge of Jesus becomes far more pressing. 
 Mike Duncan 94 churches he started under his prescriptions, likewise, scholars must read with this in mind. Holmes’s dog was not a human with a human’s complexi- ties, but the detective still had to judge its behavior—he as- sumed it was a consistent and reliable guard of the stables, and made assumptions based on this judgment. And, of course, Doyle allowed Holmes to solve his mystery through authorial fiat—we as readers know Holmes was correct in his judgment of the dog. But what about real-world AFS that cannot be con- firmed and must exist simultaneously with counterarguments that also cannot be confirmed? It is not clear which Paul is the best version, or if any of the proposed versions are even close – and even if we did know which Paul that we have in the letters, we still wouldn’t know with confidence how much of the later gospel material he had, if any. In spite of this limitation, however, the AFS in the Pauline scenario still can serve as a stepping-stone topos for proceeding with interpretations. Its “minimal hypothesis” helps reveal and organize the multitude of possibilities inherent in the question under consideration. For example, consider sub-claim (A) of the counterargument, that Paul could have considered the Jesus ma- terial that he did not mention to be trivial, well known, and not worthy of mention. Assuming Paul had visited, or would visit, each city and church that he addresses in the extent authentic letters, he would have had ample time to communicate and share traditions or sayings of Jesus with the congregations. As such, the letters are optional rather than necessary sites of Jesus mate- rial. However, the AFS demands that this claim must be recon- ciled with the extent references in the letters. If Paul felt the congregation in Corinth knew the six post-resurrection appear- ances of Jesus well, why does he bother to mention them at all, much less list them in some detail as he does? Likewise, why mention a directive of Jesus at all when his usual fallback authority throughout the epistles is Hebrew scripture? Further- more, on numerous occasions when knowledge of gospel mate- rials would have greatly aided him in argument, he does not use such knowledge.9 As such, in order to develop reasonable answers to these AFS-inspired questions, a scholar can treat the AFS as a combi- nation of a null hypothesis and stepping-stone. Unlike Holmes, 



























































 9 For example, Paul does not invoke the memory of Judas or Peter’s failure during the passion when discussing the incident at Antioch (Gal 2:11-14), and neither does he ever mention John the Baptist’s endorsement or association with Jesus to buttress his near-constant claims of Jesus’s importance. Revisiting the Argument from Silence 
 95 who proceeded in a linear fashion to the end of the mystery with the aid of his AFS, the Pauline AFS can promote the lateral and tangential motion that is suited to a hermeneutical question. 5. Conclusions The argument from silence has several qualities that make it dif- ferent from other arguments and complicate attempts to interpret its value and function. First, the AFS has an inherently vague evidential threshold for success due its reliance on negative/absent evidence, which makes both using it and evaluating it something of a subjective art. This quality leads to a demanding, non-intuitive structure for its audience, compared to arguments that do not use nega- tive/absent evidence. Second, as the “Silver Blaze” example shows, it has an investigative quality that can be used as a pathway to further ar- guments that wield greater probability and acceptability. Also, as the Pauline case shows, when the AFS enters the lawless realm of rhetoric where questions with no immanent or demon- strable correct answers exist, it becomes a dialogical topos that cannot help but attract, and develop, counterarguments. Even if its offered possibility does not pan out, as Holmes’s hunch did, an AFS can still serve an invaluable role in prompting investiga- tion. This interrogative role, however, is dependent on a will- ingness by the audience to entertain assertions of unknown probability, which in turn is dependent on the audience’s dialec- tical standards for such and their experience with the success of those standards. For example, until Holmes came along, an inside job at the stables was not under serious consideration by authorities. Wat- son or the police inspector could have insisted that Holmes con- firm the reliability of the dog and the witnesses before engaging in hypothetical musing; after all, an inside job was not a high- probability possibility to anyone else concerned with the case at that point, dog bark or no dog bark. By this methodological ob- jection, however, the good doctor would have, at the very least, delayed the successful resolution of the case. Further investiga- tion could have revealed, even, that the dog did bark, and have the unfortunate effect of sundering Holmes’s confidence in his “chain of reasoning” before it led him to stronger evidence.10 Likewise, discouragement of the AFS on the Pauline question 



























































 10 Doyle’s Watson generally reserved his criticisms of Holmes’s reasoning during a case, leaving the displays of incredulity to clients and the authorities. Mike Duncan 96 misses the opportunity to strengthen other explanations for Pauline silence. Accordingly, the most recent book collection fully concerned with the question of Pauline silence tends to re- turns to the AFS again and again in each essay, using it as a starting point regardless of the conclusions of each chapter (see Wedderburn, 1989). Third, and perhaps most importantly, however, is the AFS’s qualitative nature: its heavy dependence on the rhetor’s expert judgment and the audience’s evaluation of that judgment, which shifts the burden of proof from the stated evidence. The most popular metaphor for weighing AFS evidence, the percep- tion of auditory volume—convincing silences are often de- scribed, counter-intuitively, as “deafening”—reinforces this, not only by lending the silence a metaphorical sound, but a “deafen- ing” one, just in case no one else hears it. As such, interpretation of the strength or weakness of an AFS should proceed case-by- case, taking the acceptability issue of the relationship between the rhetor’s apparent expertise and the audience’s receptiveness into special account. Without such dialectical controls, it is eas- ier to reject an AFS out of hand for using negative evidence, when closer consideration would have been more profitable for all parties. References Aristotle. (1955). On Sophistical Refutations, On Coming-to-be and Passing Away. On the Cosmos. Cambridge: Loeb. Brickman, E. (2007). Guide to Research in Educational History (Reprint of Guide to Research in Educational History, New York University Bookstore, New York, 1949). Brickman Press. Diderjean, A. and F. Gibet. (2008). “Sherlock Holmes—An Ex- pert’s View of Expertise.” British Journal of Psychology, 99(1): 109-125. Doyle, A.C. (1967). “Silver Blaze.” In W.S. Baring-Gould (Ed.), The Annotated Sherlock Holmes, Vol II, pp. 261-281. New York: Clarkson N. Potter. Doyle, A.C. (1967). “The Adventure of the Devil’s Foot.” In W.S. Baring-Gould (Ed.), The Annotated Sherlock Holmes, Vol II, pp. 508-526. New York: Clarkson N. Potter. Eemeren, F.H. van, B. Garssen and B. Meuffels. (2009). Falla- cies and Judgments of Reasonableness: Empirical Research Concerning the Pragma-Dialectical Discussion Rules. New York: Springer. Revisiting the Argument from Silence 
 97 Garraghan, G.J., S.J. (1946). A Guide to Historical Method. J. Delanglez, S.J. (Ed.). New York: Fordham University Press. Glenn, Cheryl. (2004). Unspoken: A Rhetoric of Silence. Car- bondale, IL: Southern Illinois University Press. Hamblin, C.L. (2004). Fallacies. Newport News: Vale Press. (Reprint of 1970 edition.) Hume, D. (1993). An Enquiry Concerning Human Understand- ing. 2nd ed. Eric Steinberg (Ed.). Indianapolis: Hackett. Johnson, R.H. and J.A. Blair. (2006). Logical Self-Defense. New York: IDEA. Lange, J. (1966). “The Argument From Silence.” History and Theory, 5(3): 288-301. Stephens, C. (2011). “A Bayesian Approach to Absent Evidence Reasoning.” Informal Logic, 31(1): 56-65. Walter, N. (1989). “Paul and the Early Christian Jesus- Tradition.” In A.J.M. Wedderburn (Ed.), Paul and Jesus: Collected Essays, pp. 51-80. Worcester: Sheffield Academic Press. Walton, D. (1999). “The Appeal to Ignorance, or Argumentum Ad Ignorantiam.” Argumentation 13: 367-377. Wedderburn, A.J.M. (Ed.). (1989). Paul and Jesus: Collected Essays. Worcester: Sheffield Academic Press.